Metabolic diseases present particular difficulty for clinicians because they are often present for years before becoming clinically apparent. Clinical risk predictors of future diabetes mellitus (DM) are imperfect. A robust set of predictors of at risk individuals is of particular importance because the delay or prevention of type 2 DM may be possible via both behavioral and pharmacological approaches. One new avenue for the identification of novel disease markers is being opened by the global analysis of the human metabolome. An emerging set of technologies, based on mass spectrometry, enables the monitoring of hundreds of metabolites from biological samples. A second new avenue for the identification of novel risk factors is afforded by genome- wide association studies (GWAS), which have begun to yield robust, reproducible genetic associations with type 2 DM. Many of these genetic variants have occurred in heretofore unsuspected pathways. Indeed, many variants are not correlated with intermediate glycemic traits such as fasting glucose or insulin. Thus, there is an urgent need to characterize the metabolic consequences of these newly discovered genetic variants, and to identify additional genetic variants that are more closely related to pathogenic metabolic signatures. We postulate that combining metabolomic, genetic, and clinical data in human populations will provide a rich opportunity to identify metabolite signatures of those destined to develop overt DM. To achieve this goal, we will leverage the unique resources of the Framingham Heart Study (FHS), a well- characterized, prospective cohort in which GWAS and comprehensive metabolomic profiling are possible. In Specific Aim 1, we will document changes in plasma metabolite levels with glucose loading in individuals with and without insulin resistance. Building upon preliminary studies already performed, we will profile 500 plasma metabolites in samples obtained from ~3100 FHS subjects before and after an oral glucose tolerance test (OGTT). We will then assess the relation of two phenotypes, insulin resistance and impaired glucose tolerance, with the change in plasma metabolite levels in response to OGTT. In Specific Aim 2, we will determine whether changes in plasma metabolite concentrations with glucose loading predict the development of insulin resistance and diabetes. We will use multivariable regression to examine the relation between plasma metabolites and 2 endpoints: incident DM and insulin resistance. In Specific Aim 3, we will characterize the genetic determinants of metabolites associated with insulin resistance and diabetes. We will (A) analyze the relation of metabolites identified in Aims 1 and 2 with common genetic variants using GWAS, and (B) characterize differences in plasma metabolite concentrations in individuals with and without validated genetic polymorphisms associated with DM. Thus, our goal is to identify novel markers of preclinical disease and illuminate pathways contributing to DM. PUBLIC HEALTH RELEVANCE: Current treatments for diabetes (DM) are only partially successful, in part because they are based on limited knowledge of its root causes. Furthermore, there is no way to accurately predict who will develop DM, thus limiting our ability to intervene effectively. Our goal is to use genetics and other systematic approaches to develop novel markers of preclinical disease and illuminate our understanding of the underlying disease mechanisms in DM. [unreadable] [unreadable] [unreadable] [unreadable]